This is my 1969 MGB which I have now converted to a pure electric vehicle.

Warren Winovich, Charter Member of the Electric Auto Association, Silicon Valley Chapter kindly presented an engineering analysis of the expected performance of this electric MG. Warren is an aerospace engineer and his analysis is thorough and careful. Anyone interested in building (or converting) an electric powered car would do well to read this report.

The project comprised removal of engine and gearbox and mounting the electric motor on top of the back axle. The motor is connected to the differential of the axle using sprockets and a belt. A considerable time was spent calculating the correct sprocket combination given the motor specification, weight of the car, acceleration profile required and hill climbing ability. The gearing of the final differential was left at the standard MG factory setting - namely 4:1.

Battery technology used is Lead Acid simply because they are the cheapest. The pack will probably last two years and then, hopefully, Lithium or better technologies will be affordable. The batteries came from BobCat Batteries in San Jose - used from a telco UPS system. It's unlikely they were ever discharged. They are Sprinter AGM S12V370s - about 100AHs each. There are eight of them for 96 volts. The chargers are smart charges from Vector (bought at PepBoys) and are perfect because they offer a dedicated AGM charing profile and can supply three different charging currents - 2, 10 and 15 amps. Charging at 10Amps allows a single 110V 15A household supply to be used. If two seperate circuits are available then I can charge at 15 Amps. Also, with these smart chargers, I do not have to disconnect the batteries from each other during charging: the negative charging contacts are not gounded so they will charge the battery regardless of the absolute potential of the terminals.

The motor was supplied from High Performance Golf Cars in Ontario, California (east of LA). It is a 3-phase AC induction motor running at 96 volts. The motor and controller came as a matched pair in the sense that the controller was programmed to the characteristics of the motor. High Performance Golf Cars manufacture the motor themselves. The one I have is rated at 35HP max, 10HP continuous. I wont be burning rubber off the line but I'll be able to get up to highway speeds without difficulty.

Costs :
  • Donor car : $4000
  • Motor + controller : $4000
  • Batteries : $65 x 8 = $520 <- yes, that was a deal
  • Chargers : $60 x 8 = $480
  • Bits and pieces : $500 or so?

  • Help me finance THIS project (saving up for Lithiums!) by clicking on one of the google ads above :)

    If you're interested in the spreadsheet I created to work out the gearing and power/torque characteristics you can click on this link : Power/Torque calculations

    First things first, secure a cool number plate:)

    The motor, being basically a cylinder, was very difficult to mount. The red painted metal form was the only thing I had made for me. (I dont have the metal bending tools.) Although difficult to see, there is a 4inch hole cut in the front of the red metal form and the front part of the motor rests on that - the four bolts - threading into the aluminum motor housing are not taking the 70lb weight of the motor. The back of the motor is supported with four bolts and the whole thing rests on two parallel 2x1 rectangular tubes with orthogonal slots cut to allow for adjustment, belt tensioning and alignment.

    You can see that I had to cut away a bit of the panelling in the car. I'm keeping an eye on the rigidity of the car. I might have to add some strengthening. Note that the motor, mounted to the axle, is unsprung - meaning that the motor bounces up and down relative to the body of the car. This is quite amusing to watch as you drive along.

    No gear lever. No need for a clutch. My speed-o is a hand-held GPS unit (honestly, officer!) and I have two safety items - a fire extinguisher and a pair of cable cutters. It would be nice to have some carpets too.

    By tightening this bolt and nut combo, I can adjust the tension in the belt.

    The motor specs.

    This is the bottom sprocket. The sprocket mounts to a 2 inch diameter shaft which bolts to the original flange which is the input to the original differential. The metal plate with the irregularly cut hole was going to be welded to the diff housing to act as a support for the weight of the motor but now it's really just cosmetic and acts as a fail-safe if the welds on the axle fail. The main reason for not welding it is that it's very difficult to successfully weld to cast iron. All sorts of horrible spiderweb-like cracks can occur and a broken diff housing is not something I wanted. One really time-consuming task was to make sure that the shaft rotated perfectly concentrically with the diff input.

    I was worried about the canterlever effect of the motor hanging out on the framework away from the centerline of the axle. All the weight of the motoris actually being supported by the welds on the axle. But metal is strong and hopefully my welding is good enough.

    When I connected up the three cables from the controller to the motor I realized that the gap between the connectors was really small. I was worried, with the motor bouncing up and down with the wheels that the relative movement of the cables and the motor could cause a short. Hence the elaborate perspex enclousure plus dividers. (This took all of 15 minutes to make.)

    No gas tank and hence no filler cap. I might be putting a speaker in the hole to play standard gas engine sounds from am MP3 player. (I previously recorded the sound of the engine before removing it :) )

    It's funny how quickly wiring can look complicated but really it's all pretty straight-forward. The controller is the black box mounted to a piece of aluminum bolted to the panel behind the driver. It all fitted rather well and conveniently.

    I found a copper guttering company (Kobett Metals) who fortunately had perfectly cut lengths of solid copper bar which I gratefully accepted (and paid for). Thanks George!

    Here is the Curtis controller, the 400Amp fuse, shunt for measuring the current flow, the 'contactor' (a relay) (the white cylinder unit) and a bunch of copper bar connections. Being a bit concerned about the proximity of the positive and negative cables, I covered negative supply with extra tubing. A short would not be good - I'd be using my fire extinguisher if contact was made. The perspex cover is there to stop accidental shorts when tightening bolts etc.

    More perspex protection over the battery terminals. The thick cable is 2/0 welding cable and the thinner black cable is for the chargers. The thinnest cable is for the instrumentation (measuring the voltage of the batteries, that is).

    I was fortunate that there are some so-called smart chargers on the market now. These are computer controlled and have the correct charging profile for AGM batteries. Not only that but the terminals are floating meaning that they don't care what the voltage is on either the positive or negative terminal with respect to ground. All they care about is the relative voltage across the terminals. This is excellent news because I can charge all the batteries without having to disconnect anything and it's very much preferable to charge each battery individually rather than in series.

    Here's the front of the car - you should have seen the expression on the face of the DMV inspector when I opened the hood:). For those paying attention you'll notice that there are six chargers and only five batteries. The sixth battery is hidden below. There are two batteries in the back. This 6 and two arrangement made the ride height of the car pretty much the same as it was originally. The batteries are secured at their base with metal brackets and then strapped down with roof-rack straps. The arrangement seems to work.

    I dont have a dc to dc converter, just a seperate small battery for the lights and horn. There's no power steering or power brakes on this car and certainly no AC. So there's little need for power for these accessories. In this picture you can also see the accelerator device on the bulkhead.

    This is under the car at the front. The two aluminum frames support just one battery. My original plan was to place four batteries under the car in the transmission well (and hence the substantial support structure). However, in the end I just put one down there but kept the frames as a useful support for all the cabling that needed to go from the front to the back of the car.

    Here's a close up of the battery to battery connection. The 2/0 welding cable is squeezed tightly between two 3/16inch thickness bar which bends nicely to accomodate the wire and form an excellent, non-moving contact. The thinner black wire is the charger wire.

    I think I was lucky to get a car that had reasonable bodywork. There are plenty of holes in the floor pans but this is a British car and they were probably manufactured with rusty metal in the first place:) I have no intention of driving this car in the wet.

    Here's the one battery at the front, underneath. All the brackets are aluminum - thanks to Sims Metal recyclers for the supply.

    The white bracket to which the three motor cables are clamped is quite substantial. I didn't like the fact that the cable was moving at the copper connector end on the controller. (Remember that the motor goes up and down relative to the controller and the rest of the car. So I clamped the cables solidly to the car and protected them with bits of garden hose. The two lower motor cable could hit the red metal frame so I covered those too with some old rubber hose.

    This is the back of the diff. (todo - add the missing bolts and replace the original diff oil with something synthetic and less viscous. The diff does get warm when driving.) Note also the two 3x1 rectangular tube sections welded on to the axle. I was fortunate that this car had a steel axle - not a cast version.

    You can see the motor support structure again here - also note the shims under the red motor frame. It was difficult to get the motor and differential shaft to line up perfectly. I still don't think it perfect.

    The white square tube is actually a piece of plastic garden fencing from Home Depot. It was perfect to feed the thick cables through and past all the bits of moving metal comprising the motor mounting and the axle.

    free counters